Characterizing the Google Books Corpus: Strong Limits to Inferences of Socio-Cultural and Linguistic Evolution
By Eitan Adam Pechenick, Christopher M. Danforth, Peter Sheridan Dodds. 2015
329 citations recorded by the source corpus.
It is tempting to treat frequency trends from the Google Books data sets as indicators of the “true” popularity of various words and phrases. Doing so allows us to draw quantitatively strong conclusions about the evolution of cultural perception of a given topic, such as time or gender. However, the Google Books corpus suffers from a number of limitations which make it an obscure mask of cultural popularity. A primary issue is that the corpus is in effect a library, containing one of each book. A single, prolific author is thereby able to noticeably insert new phrases into the Google Books lexicon, whether the author is widely read or not. With this understood, the Google Books corpus remains an important data set to be considered more lexicon-like than text-like. Here, we show that a distinct problematic feature arises from the inclusion of scientific texts, which have become an increasingly substantive portion of the corpus throughout the 1900s. The result is a surge of phrases typical to academic articles but less common in general, such as references to time in the form of citations. We use information theoretic methods to highlight these dynamics by examining and comparing major contributions via a divergence measure of English data sets between decades in the period 1800–2000. We find that only the English Fiction data set from the second version of the corpus is not heavily affected by professional texts. Overall, our findings call into question the vast majority of existing claims drawn from the Google Books corpus, and point to the need to fully characterize the dynamics of the corpus before using these data sets to draw broad conclusions about cultural and linguistic evolution.
Subfields: Cultural Studies, Sociology and Political Science, Computational Linguistics & Semantics, Social Sciences & Humanities
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Cultural Studies
- Dong Nguyen
- Franciska de Jong
- Jure Leskovec
- Dan Jurafsky
- A. Seza Doğruöz
- Momin M. Malik
- William L. Hamilton
- Stephan Mandt
In Sociology and Political Science
- Dong Nguyen
- William L. Hamilton
- Dan Jurafsky
- James A. Evans
- Jure Leskovec
- Momin M. Malik
- Matt Taddy
- Carolyn Penstein Rosé
In Computational Linguistics & Semantics
In Social Sciences & Humanities
- Franciska de Jong
- Dong Nguyen
- Jure Leskovec
- Dan Jurafsky
- A. Seza Doğruöz
- Momin M. Malik
- Matt Taddy
- William L. Hamilton
Citations in the corpus
- Computational Sociolinguistics: A Survey (2016, background): “Some studies have therefore treated the use of hashtags and user mentions as proxies for the target audience (Nguyen, Trieschnigg, and Cornips 2015; Pavalanathan and Eisenstein 2015a).”
- A Hierarchy of Limitations in Machine Learning (2020, background): “Such measurement problems are over and above problems coming from an unclear sampling frame (Pechenick et al., 2015).”
- Inducing Domain-Specific Sentiment Lexicons from Unlabeled Corpora (2016, methodology): “We constructed lexicons from COHA, since it was carefully constructed to be genre balanced (e.g., compared to the Google N-Grams; Pechenick et al., 2015).”
- The Geometry of Culture: Analyzing the Meanings of Class through Word Embeddings (2018, background): “Google Ngrams have been subject to criticism because the composition of the corpus in a given year may not be representative of total literary output (Pechenick, Danforth, and Dodds 2015).”